Performance Analysis of UPQC using Mittag Leffler Polynomial based Neural Network Control

Abstract

This paper investigates the performance of Unified Power Quality Conditioner (UPQC), which comprises of shunt and series voltage source converter (VSC) under voltage Sag, voltage swell, and load unbalancing. UPQC is integrated with 25 kW power rated Photo Voltaic (PV) source. A hybrid control technique is executed for UPQC. The Mittag Leffler Polynomial based Neural Network (MiLeP) control is executed for the generation of switching signals to shunt VSC and ‘d-q' with MiLeP filter control for the generation of gating signals to series VSC. The shunt compensator is developed to mitigate current related Power Quality (PQ) issues, and the series compensator is developed to alleviate voltage related PQ issues. The THD of the source current is below 5% which is under the IEEE-1547 norms. The load voltage is maintained at pre-sag and pre-swell conditions under both Voltage Sag and Voltage swell. In MATLAB Simulink, the simulation is executed and the results are examined with the proposed control algorithm. The results show improved performance of PV- UPQC under both steady-state and dynamic operation.